{"ID":6620429,"CreatedAt":"2026-07-15T01:01:48.440468303Z","UpdatedAt":"2026-07-17T01:19:05.081438488Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2607.12248","arxiv_id":"2607.12248","title":"When Directional Accuracy Lies: A Base-Rate-Honest Benchmark for LoRA-Adapted TimesFM on Equity Forecasting","abstract":"Large pretrained time-series models such as TimesFM are attractive for financial forecasting, but raw directional accuracy is a misleading scoreboard in equity markets. An early LoRA adapter in this project appeared to reach roughly 80% directional accuracy; we show this is not evidence of skill. Over a long horizon in a rising market, a trivial \"always-up\" rule attains comparably high accuracy without using the input at all. To separate genuine skill from this base-rate artifact, we build a reproducible, frozen-data benchmark with expanding walk-forward folds, a stratified held-out-ticker split, honest baselines (zero-shot TimesFM, always-up, random-walk, persistence, AR(1)), and paired significance tests (McNemar, Diebold-Mariano) under Benjamini-Hochberg FDR control. We apply the identical method to two universes -- a tech-heavy NASDAQ-100 and a broad S\u0026P 500 -- reporting excess accuracy over the always-up base rate. Three findings replicate. First, when the historical ~80% condition is recreated, the high number is a base rate of ~0.70 that the fine-tuned model scores below. Second, pooled LoRA shows no directional skill over the base rate at any horizon on either universe (negative at the six-month horizon). Third, per-sector specialization is significantly worse than a single pooled adapter (Diebold-Mariano p\u003c0.001 on held-out stocks at h=128). Fine-tuning's only measurable benefit is a statistically significant reduction in point-forecast error relative to zero-shot TimesFM, which nonetheless does not beat naive baselines and confers no tradeable directional edge. The contribution is methodological: a defensible, fully seeded protocol that prevents the base-rate trap, together with the replicated negative result it produces.","short_abstract":"Large pretrained time-series models such as TimesFM are attractive for financial forecasting, but raw directional accuracy is a misleading scoreboard in equity markets. An early LoRA adapter in this project appeared to reach roughly 80% directional accuracy; we show this is not evidence of skill. Over a long horizon in...","url_abs":"https://arxiv.org/abs/2607.12248","url_pdf":"https://arxiv.org/pdf/2607.12248v1","authors":"[\"Taizhen Cheung\",\"SA Kwon\"]","published":"2026-07-14T01:23:51Z","proceeding":"q-fin.ST","tasks":"[\"q-fin.ST\",\"cs.LG\"]","methods":"[\"LoRA\"]","has_code":false}